Optimization of falling film thermosyphons bundle arrangement for large-scale cooling applications by genetic algorithm

Abstract

Two-phase traditional thermosyphons (TTS) are passive heat exchangers with low thermal resistance for heat transfer. An advanced thermosyphon system named falling-film thermosyphon (FFTS) is developed and introduced in this paper, which eliminates several limits of TTS and improves heat transfer performance. FFTS applications are preferred for large-scale heat transfer applications. In this work, a tube bank of FFTS modules is integrated into a power plant to condense water vapor of the flue gas. Axial- and cross- flow are both analyzed with inline and staggered arrangements. The numerical simulation of vapor condensation in the flue gas is developed, and the results are compared to previously published experimental data. The accuracy of model is validated. The Non-dominated Sorting Genetic Algorithm, version 2 (NSGA-II) is implemented to improve the thermal performance with low material consumption and pressure drop for tube banks. Several parameters of the configuration are optimized. As a result, smaller diameter tubes with longer length can gain better heat transfer performance, however, resulting in a higher pressure drop. The optimized arrangements can improve 60–70% condensation rate and reduce up to 80% material consumption with allowable pressure loss compared to a baseline geometry.